Complications of Poor Cervical Alignment in Patients Undergoing Posterior Cervicothoracic Laminectomy and Fusion

Original Article

Complications of Poor Cervical Alignment in Patients Undergoing Posterior Cervicothoracic Laminectomy and Fusion Brooke T. Kennamer1, Marc S. Arginteanu1,2, Frank M. Moore1,2, Alfred A. Steinberger1,2, Kevin C. Yao1,2, Yakov Gologorsky1,2

OBJECTIVE: This study sought to determine whether a relationship exists between caudal instrumented level and revision rates, neck disability index scores, and cervical alignment in patients undergoing multilevel posterior cervical fusion.

CONCLUSIONS: This study suggests that constructs terminating in the proximal thoracic spine have similar revision rates, postoperative neck disability index scores, and radiographic measurements as those terminating in the cervical spine. Poor cervical alignment, as evidenced by increased sagittal vertical axis, cervical kyphosis and T1 slope, predicts need for revision and of poorer clinical outcomes.

-

-

METHODS: This study examined a dataset of all patients undergoing posterior cervical decompression and fusion at ‡3 levels, terminating between C4 and T4, between January 2010 and December 2015, with at least 12 months of clinical follow-up. Patients were separated into cohorts based on caudal level of the fusion: C6 (or more cranial), C7, T1, or T2 (or more caudal). Revision rate, neck disability index score, sagittal vertical axis, T1 slope, and cervical lordosis were recorded. Linear regression and multivariate analysis were performed to identify independent predictors of patient outcomes and disparities between ending constructs in the cervical and the thoracic spine.

INTRODUCTION

-

RESULTS: The overall revision rate was 10.8% (n [ 24). No statistically significant difference in the revision rate was identified between fusions terminating at C6 or cranial, C7, T1, or T2 and caudal (P [ 0.74). Revision correlated strongly with increased sagittal vertical axis (P [ 0.002) and T1 slope (P [ 0.04). Increased neck disability index score correlated with revision rate (P [ 0.01), cervical kyphosis (P < 0.001), and increased sagittal vertical axis (P [ 0.04).

-

Key words Cervical kyphosis - Cervicothoracic fusion - Cervicothoracic junction - Multilevel posterior cervical fusion - Posterior cervical fusion - Sagittal vertical axis -

Abbreviations and Acronyms ACDF: Anterior cervical discectomy and fusion BMI: Body mass index CL: Cervical lordosis HRQOL: Health-related quality of life NDI: Neck Disability Index ODI: Oswestry Disability Index PCDF: Posterior cervical decompression and fusion

WORLD NEUROSURGERY -: e1-e7, - 2018

P

atients with cervical spondylotic myelopathy, multilevel cervical stenosis, and cervical deformity frequently undergo posterior cervical decompression and fusion.1,2 In similar clinical scenarios and radiographic findings, some surgeons prefer to terminate the construct in the cervical spine, whereas others elect to cross into the proximal thoracic spine. Although a substantial amount of literature details the challenges of performing surgery at the cervicothoracic junction,3,4 there is a scarcity of literature to assist the surgeon in determining the most appropriate caudal level to terminate the fusion. Furthermore, recommendations for caudal level of a posterior cervical decompression and fusion (PCDF) remain variable and debated.5-7 The cervicothoracic junction has several unique biomechanical and structural characteristics. The cervical spine is lordotic, whereas the thoracic spine is kyphotic. The cervical spine has a greater degree of mobility in flexion, extension, and side-bending

SIF: Symptomatic instrumentation failure SVA: Sagittal vertical axis VAS: Visual Analog Scale From the 1Englewood Hospital and Medical Center, New Jersey; and 2Mount Sinai Medical Center, New York, New York, USA To whom correspondence should be addressed: Brooke T. Kennamer, B.A. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.10.062 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.

www.WORLDNEUROSURGERY.org

e1

ORIGINAL ARTICLE BROOKE T. KENNAMER ET AL.

CERVICAL ALIGNMENT IS CRUCIAL IN POSTERIOR CERVICOTHORACIC FUSION

than the thoracic spine owing to articulation of the ribs and osteoligamentous structures in the thoracic spine.3,8 The cervical spine is responsible for maintaining the head over the body and for as maintaining horizontal gaze. The thoracic kyphosis enables lung expansion. Previous studies have aimed to guide the surgeon in selecting the most appropriate caudal level of fusion, but they have reached mixed results.5-7 The purpose of this study was to determine whether there are relationships between revision rates, Neck Disability Index (NDI) scores, and cervical alignment in patients undergoing multilevel posterior cervical fusion terminating at C6 (or cranial), C7, T1, or T2 (or caudal). MATERIAL AND METHODS After institutional review board approval (EH IRB 4777) was obtained, all adult patients undergoing
3 level PCDF at our practice between January 2010 and December 2015 were identified. Only fusions terminating in the cervical or proximal thoracic spine were included (up to T4). Patients with less than 12 months of follow-up were excluded. Patients were also excluded if surgery was performed for tumor, trauma, or infection. Patients were also excluded if surgery required 3-column osteotomy for advanced deformity. The records draw from a single practice with 5 neurosurgeons, 2 of whom are fellowship trained in spine. All of the

operations evaluated were performed by at least 1 of the 5 neurosurgeons at this practice. This analysis included the retrospective review of prospectively defined data. Several patient-related variables were collected: age, body mass index (BMI), gender, active smoking status, preoperative diagnosis, existence of any other cervical operations (prior or simultaneous anterior cervical discectomy and fusion), and existence of any other spine operation performed by this practice or elsewhere. Patients were separated into 4 cohorts based on the caudal level of the fusion: C6 (or cranial), C7, T1, or T2 (or caudal). Operative variables collected included levels decompressed, cranial and caudal level of fusion, length of construct, screw type (i.e., lateral mass, pars, pedicle), and hardware manufacturer. Postoperative variables included all complications from surgery (i.e., acute or late postoperative infection, adjacent segment disease, and pseudarthrosis), need for revision (both patients who had been offered additional surgery and those who underwent additional surgery), reason for revision, levels decompressed or fused during the revision, radiographic measurements (sagittal vertical axis [SVA], cervical lordosis, and T1 slope), NDI scores, and months of follow-up. Adjacent segment disease was defined as radiographic disc degeneration precipitating worsening surgical outcome, sometimes necessitating additional surgery. Radiographic measurements for this practice are obtained preoperatively, and then postoperatively at 3 weeks, 3 months, 6 months,

Table 1. Demographic Information and Clinical Variables Caudal Level N (N¼221) Age (years) Gender (F/M) Active Smoker, N¼25

C6 (or more Cranial)

C7

T1

T2 (or more Caudal)

36

140

41

4

P-Value

67.67 (45e88)

65.3 (37e91)

68.7 (44e86)

62.5 (46e74)

0.80

47.2%/52.8% 17F/19M

44.3%/55.7% 62F/78M

43.9%/56.1% 18F/23M

0%/100% 0F/4M

0.90

3

14

7

1

0.78

29.7 (18.5e46)

28.8 (18.4e42.6)

28 (18.1e44.9)

28.4 (21.5e38)

0.44

Average length of construct

3

3.85

4.9

7

0.7

Months follow-up (months)

51.2 (12e96)

50.1 (12e94)

51.9 (12e98)

55.5 (36e88)

0.50

4

25

11

2

0.44

Stenosis with Myelopathy, N¼93

14

61

17

1

0.83

Stenosis without Myelopathy, N¼128

22

79

24

3

0.90

Pseudarthrosis, N¼1

0

1

0

0

0.92

Spinal Deformity, N¼15

0

11

4

0

0.89

C1

0

1

0

0

1

C2

1

7

3

2

1

C3

35

102

31

2

0.2

C4

0

30

4

0

0.07

C5

0

0

3

0

1

BMI (points)

Anterior support (before or simultaneous, N¼42) Pre-op Diagnosis

Cephalad level

BMI, body mass index.

e2

www.SCIENCEDIRECT.com

WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.10.062

ORIGINAL ARTICLE BROOKE T. KENNAMER ET AL.

CERVICAL ALIGNMENT IS CRUCIAL IN POSTERIOR CERVICOTHORACIC FUSION

1 year, and yearly thereafter. Cervical lordosis, SVA, and T1 slope were measured. All the radiographic measurements were obtained by 1 reviewer. All the radiographic measurements were verified by an additional reviewer. Cervical lordosis measurement was obtained by use of the Cobb angle technique from C2 to C7. Although the Harrison posterior tangent line method has been suggested to be the most accurate method for measuring cervical lordosis,9 we used the Cobb angle method because is widely accepted.10 It is notable that using the C2eC7 Cobb angle may underestimate cervical lordosis in comparison with the Harrison technique. The C2eC7 SVA was calculated regionally by measuring the horizontal distance between the posterosuperior endplate of C7 and a plumb line dropped from the centroid of C2.10 The T1 slope was measured as the angle between the superior endplate of T1 and a horizontal line. Although T1 slope can be difficult to measure because of the radiopacity of soft tissue shadows like the shoulders at the cervicalethoracic junction, the brightness was adjusted to aid the ability to take a measurement. The measurements were tested as continuous variables. In addition to office evaluations, all patients were called for long-term follow-up. The patients were asked if they had any new radiographic data since their last office visit or additional cervical spine operations that the practice was not aware of. All patients had an NDI score calculated at the time of their most recent inoffice follow-up visit. Patients were excluded from the study if there was less than 12 months of clinical follow-up time. Statistical Analysis The patient data were stratified into 1 of the 4 cohorts, and these groups were compared. The Fisher exact test was used for comparing binary or categoric variables among groups. The Kruskal-Wallis test was used to compare continuous variables such as age, months of follow-up, and BMI. A Bonferroni correction was applied to account for multiple dependent and independent statistical tests. All statistical analysis was done with Microsoft Excel and R v. 3.3.3 (R Foundation for Statistical Computing, Vienna, Austria). The confidence interval was 95%, and statistically significant calculations were defined as a <0.05. After bivariate linear regression was completed, multivariate analysis was computed, and multiple models were created in R v. 3.3.3. Multivariate analysis was computed with caudal level as the dependent variable with several independent variables summarized in Table 1. Another model used revision rate as the dependent variable and another used the NDI score as the dependent variable. There were several independent variables, including age, sex, BMI, number of levels fused, cranial level, caudal level, SVA, cervical lordosis, T1 slope, and anterior support. This analysis is summarized in Table 2. Last, the reason for revision was studied to identify whether there was a statistically significant difference in reasons for revision between different cohorts. RESULTS After the removal of patients who did not meet the inclusion criteria, 221 patients were identified. Radiographic data were available for 177 (79.7%) patients. In addition to standard office follow-up, long-term telephone follow-up was achieved for 138

WORLD NEUROSURGERY -: e1-e7, - 2018

Table 2. Predictors of Revision and Neck Disability Index (NDI) Scores Variable

P Value for Revision

P Value for NDI

Male vs.female

0.78

0.001

Age (per year)

0.14

0.92

BMI (per point)

0.84

0.47

Active smoker

0.99

0.87

Anterior support

0.34

0.86

Length of construct

0.99

0.89

Cephalad level C1

0.97

0.79

Cephalad level C2

0.92

0.41

Cephalad level C3

0.73

0.40

Cephalad level C4

0.42

0.49

Caudal level C5

NA

0.83

Caudal level C6

0.96

0.86

Caudal level C7

0.97

0.95

Caudal level T1

0.82

0.87

Sagittal vertical axis

0.002

0.04

Cervical lordosis

0.53

<0.001

T1 slope

0.04

0.71

Worker’s compensation/MVC

NA

<0.001

Postoperatively infection

NA

0.02

Revision

NA

0.0

BMI, body mass index; MVC, motor vehicle collision.

(62.4%) patients. The average follow-up time was 50.7 months (range, 12e98 months). Thirty-six patients had constructs terminating at C6 or cranial; 140 patients had a caudal level at C7, 41 patients had a caudal level at T1, and 4 patients had constructs terminating at T2 or caudal. The median length of the constructs was 4 levels (i.e., C3eC7). There was no statistical difference in age, BMI, sex, anterior support, months of follow-up, preoperative diagnosis, smoking status, cervical lordosis, SVA, and revision rate between the 4 cohorts (Table 1 and Table 2). Anterior support was defined as an anterior cervical discectomy and fusion (ACDF) before the PCDF or an ACDF simultaneously performed during the PCDF being studied. Incidentally, preoperative diagnoses exceed the total number because some patients had multiple in-office preoperative diagnoses (for example, pseudarthrosis after anterior cervical discectomy, or as stenosis with radiculopathy). All the results summarized in the tables were obtained from multivariate analysis. Table 2 summarizes the results of multivariate analysis of predictors of postoperative NDI scores and need for revision. There was a statistically significant relationship between revision rate and increasing SVA (P ¼ 0.006) and higher T1 slope (P ¼ 0.04). The average SVA of patients requiring revision was 55.9 mm, compared

www.WORLDNEUROSURGERY.org

e3

ORIGINAL ARTICLE BROOKE T. KENNAMER ET AL.

CERVICAL ALIGNMENT IS CRUCIAL IN POSTERIOR CERVICOTHORACIC FUSION

Table 3. Reasons for Revision Caudal Level Revisions total (n ¼ 24) Revision rate (10.8%)

C6 (or Cranial)

C7

T1

T2 (or Caudal)

P Value

2

14

8

0

0.74

5.6%

10.0%

19.5%

0.0%

0.74

Adjacent level disease (n ¼ 13)

2

8

3

0

0.73

Nonunion/hardware failure (n ¼ 10)

0

5

5

0

0.17

Spinal deformity (n ¼ 2)

0

2

0

0

1

with 43.5 mm for those who did not. Increased NDI score correlated with need for revision (P ¼ 0.01), cervical kyphosis (P < 0.001), increased SVA (P ¼ 0.04), and worker’s compensation/motor vehicle accident patients (P < 0.001). The overall revision rate was 10.8% (n ¼ 24). The revision rates for the 4 cohorts were 5.6%, 10.0%, 19.5%, and 0%, respectively. There was no statistically significant relationship between caudal construct level and revision rate (P ¼ 0.74). Of note, only 4 patients were in the T2 or caudal cohort, precluding the establishment of any statistically significant conclusions. Furthermore, this group was isolated from the T1 cohort in an effort to draw a meaningful comparison between the T1 cohort in this study and a T1 cohort in other studies. Construct length did not correlate with need for revision (P ¼ 0.99). Six constructs failed cranially, and 19 constructs failed caudally. One patient’s fusion failed both cranially and caudally. There were notable differences in the reason for revision surgery between the 4 groups (Table 3). Nonunion/hardware failure was a more common reason for revision in constructs terminating at T1, whereas adjacent level degeneration (primarily at C7eT1) was a more common reason at C7. Despite these differences, there was no statistical difference between reason for revision and caudal termination level, likely owing to low incidence. Notably, 1 patient had both adjacent segment disease and hardware failure, so the total number exceeds the total revisions. There was a statistically significant relationship between increasing SVA misalignment and adjacent segment disease (P ¼ 0.04). Of the 24 patients requiring revision, the median SVA was 55.2 mm, as opposed to 39.4 mm in the nonrevision cohort (P ¼ 0.007). Table 4 presents the postoperative radiographic measurements obtained for each cohort and the respective cervical lordosis and SVA. Again, there was no statistically significant relationship between postoperative cervical lordosis or SVA and caudal level. There appears to be a trend in patients undergoing surgery terminating more distally to have slightly smaller cervical lordosis, and, conversely, increased SVA.

Table 5 shows the NDI scores for each cohort. The mean NDI for all patients was 22.0% (classified as mild disability). The NDI was similarly distributed among all cohorts. As noted earlier, patients with potential secondary gain, including those involved in litigation related to worker’s compensation and motor vehicle accidents, had higher NDI scores (indicating higher disability).

DISCUSSION This study does not support the finding of a more favorable outcome with constructs terminating either proximally or distally to the cervicothoracic junction. We found that the most significant predictor of disability and need for revision was poor cervical alignment. This study suggests that the notion of “most ideal level” to end a construct is misguided. The results of this study are distinct from those of similar studies such as those by Schroder et al.5 and Wagner et al.,7 in that this study does not support the finding of a more favorable outcome with constructs terminating either proximally or distally to the cervicothoracic junction. Some authors have suggested that crossing the cervicothoracic junction leads to lower rates of pseudarthrosis,6 but this notion seems counterintuitive and is not substantiated by our findings. However, extending a construct into the proximal thoracic spine increases the number of levels fused, increases neck stiffness and neck discomfort, and is associated with higher rates of blood loss, increased operative times, and lengths of stay in the hospital.6 In the same vein, Schroder et al.5 studied 219 patients who underwent multilevel posterior cervical decompression and found that fusions terminating at C7 were 2.29 times more likely to undergo a revision than were those whose constructs terminated at T1. In contradistinction, Wagner et al.7 studied 247 patients who underwent similar surgeries and found that there was an increased rate of instrumentation failure in cases that terminated distal to the cervicothoracic junction (P ¼ 0.04).

Table 4. Postoperative Radiographic Measurements Caudal Level N (n ¼ 177, 79.7%)

C6 or Cranial

C7

T1

T2 or Caudal

P Value

32 (88.9%)

109 (77.9%)

33 (80.5%)

3 (75.0%)

Postoperative cervical lordosis (degrees)

3.9 (16.5 to 40.7)

2.3 (19.5 to 25.1)

1.4 (23.4 to 22.5)

0.5 (15.4 to 11.5)

0.85

Postoperative sagittal vertical axis (mm)

37.5 (4.2 to 77.4)

45.6 (8.e98.)

48.2 (6.7e79.9)

64.0 (34.3e84.0)

0.36

e4

www.SCIENCEDIRECT.com

WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.10.062

ORIGINAL ARTICLE BROOKE T. KENNAMER ET AL.

CERVICAL ALIGNMENT IS CRUCIAL IN POSTERIOR CERVICOTHORACIC FUSION

Table 5. Neck Disability Index Scores Caudal Level

C6 or Cranial

C7

T1

T2 or Caudal

Number: 220

36

139

41

4

Mean NDI 22.0%

19.1%

22.3%

21.1%

22.0%

NDI: no disability (score <10%) n ¼ 82

15

46

19

2

NDI: mild disability (10%e28%) n ¼ 59

12

41

6

0

NDI: moderate disability (30%e48%) n ¼ 63

6

42

13

2

NDI: severe disability (>50%) n ¼ 16

3

10

3

0

NDI, Neck Disabity Index.

Our study did not find any statistical difference in reoperation rates between T1 compared with C7, respectively. Several factors may contribute to the inconsistent findings between these studies. The total revision rate for Schroeder et al.5 was 27.8%, whereas the total revision rate for this study was 10.8%. The discrepancy in revision rates may influence the results of all studies. The lower revision rate in this study may be accounted for by the surgeons’ experience and by level and patient selection. Many of the revisions in the study by Schroeder et al.5 were reported from patients by telephone and were performed at another institution; in this study, all reasons for revision were obtained and investigated. Another potential contributing factor to the discrepancy in revision rate may be accounted for by the follow-up time. The average in this study was 50.7 months, similar to that of the study by Schroeder et al.,5 which had an average follow-up time of 49.8 months. Wagner et al.,7 however, had the shortest follow-up time: 17 months. A shorter follow-up time influences the timeframe over which instrumentation failure and adjacent level disease develop, which may necessitate a revision. Based on a meta-analysis of 83 studies, the prevalence of reoperation adjacent segment disease after cervical spine surgery was 5.78% and increases by 0.24% annually.11 The development of adjacent segment disease may be related to adjacent biomechanical stress, age, segment degeneration, altered sagittal alignment, and cervical kyphosis.12,13 Although the total numbers are consistent among this study, the study by Schroeder et al.,5 and the study by Wagner et al.,7 operative caudal level is more evenly distributed in Schroeder et al.5 and Wagner et al.7 than in this study, which may bias the results, lessening the ability to draw statistically significant conclusions in the T1 and the T2 and caudal cohorts. In our patient cohort, the determination of the caudal level was made by the surgeon’s preference. There was general agreement among surgeons of the practice that all stenotic levels needed to be decompressed, and that (at least) all decompressed levels would be included in the fusion construct. Similarly, if there was moderate to severe spondylosis of the disc space immediately caudal to the proposed construct (even in the absence of stenosis at that level), the degenerated adjacent level was included in the fusion construct. Patients with severe preoperative sagittal malalignment tended to require 3-column osteotomy, but those patients were excluded from this cohort analysis. We did not screen

WORLD NEUROSURGERY -: e1-e7, - 2018

patients for low bone density preoperatively, which may have potentially influenced the results of this analysis. This study supports the suggestion that spine surgeons make operative approach decisions based on radiographic measurements, such as cervical lordosis and SVA in patients with cervical spondylotic myelopathy.10,14,15 Cervical sagittal alignment has been correlated to regional disability, general health scores, and myelopathy severity.10 This study used the NDI to quantify regional disability; this tool has been shown to be a useful scale to evaluate patients with neck pain.16 This study identifies a relationship between revision rate and increased SVA and increasing T1 slope. T1 slope helps determine the subaxial lordosis and is strongly correlated with SVA.10 Several studies support the use of T1 slope in presurgical planning and its correlation to global spinal alignment.10 Prior studies13,17,18 have suggested a threshold of 40 to 45 mm for C2eC7 SVA, above which health-related quality of life (HRQOL) outcomes are poor. This statistically significant finding was confirmed in our study as well, with SVA and NDI scores tested as continuous variables. Of the 24 patients requiring revision in our cohort, the median SVA was 55.2 mm, as opposed to 39.4 mm in the nonrevision cohort (P ¼ 0.007). We identified increased NDI to correlate with increased SVA, need for revision, cervical kyphosis, and increased T1 slope. It seems reasonable to conclude that these pooled findings suggest that patients with increased SVA have worse clinical outcomes and poorer HRQOL and NDI scores.13,17-21 However, there remained a lack of evidence supporting a relationship between cervical kyphosis and HRQOL,18 except that identified in our study. The recurrent finding is that there is increased neck pain in patients who have postoperative cervical kyphosis.22 Furthermore, these patients tend to have less neurologic improvement postoperatively.15 There is no definitive amount of ideal cervical lordosis postoperatively, and the objective tends to be to correct cervical kyphosis to as close to neutral as possible.13 Notably, there is no difference between caudal level and SVA or caudal level and cervical lordosis. This finding is consistent with Truumees et al.,6 who did not identify a difference in postoperative SVA in constructs terminating in the cervical or thoracic spine. Our hypothesis is that sagittal plane abnormalities place excess stress on surgical implants, which predisposes to hardware failure

www.WORLDNEUROSURGERY.org

e5

ORIGINAL ARTICLE BROOKE T. KENNAMER ET AL.

CERVICAL ALIGNMENT IS CRUCIAL IN POSTERIOR CERVICOTHORACIC FUSION

or osteoligamentous failure/fracture. Sagittal imbalance at the nonoperated levels may also affect failure of instrumentation.14 Additionally, long-segment fixation, smoking, old age, and fixation crossing a mobile to immobile segment are possible risk factors for symptomatic instrumentation failure.11 Although smoking is a contributing factor for pseudarthrosis,23 there was no statistically significant relationship between revision and smoking or smoking and hardware failure in our cohort. It is important to note that this study did not evaluate pathologic changes in the lower limbs, which may contribute to cervical misalignment. The cervical spine influences subjacent global spinal alignment and pelvic tilt as compensatory changes occur to maintain horizontal gaze. In a patient with global sagittal misalignment, cervical lordosis increases as a compensatory mechanism.13 This study did not evaluate other radiographic parameters, including compensatory mechanisms such as thoracic hypokyphosis, pelvis retroversion, and lumbar hyperlordosis.13 Lumbar surgeries that were performed by neurosurgeons at this practice were tracked, but this was not informative enough to draw conclusions about global sagittal alignment in these patients. Full-length 36-inch scoliosis films were not performed routinely in our practice during this time period, although they are now routinely done. There are several limitations to this study, particularly its retrospective nature, which implies inherent problems such as surgeon bias and variation. As previously mentioned, the uneven caudal distribution and small numbers in the T2 and caudal cohort may have biased the results. Specifically, although revision surgery rates were not statistically different based on lowest instrumented

REFERENCES 1. Fehlings M, Kopjar S, Yoon B, Sangwook A, Massicotte P. Anterior vs. posterior surgical approaches to treat cervical spondylotic myelopathy: Outcomes of the prospective multicenter AO Spine North America CSM study in 278 patients. Spine. 2013;38:2247-2252.

vertebra, the absolute revision surgery rates were 4 times higher in the T1 group than in the C6 group and twice as high in the T1 group as in the C7 group. This certainly may be due to the underpowerment of the study resulting from the low incidence of constructs ending at T1 overall. The NDI score obtained for patients is influenced by confounding factors that affect a patient’s quality of life, and the day the patient was questioned may have been a good or bad day for that patient. The lack of baseline NDI scores and preoperative radiographic measurements to compare postoperative values to are other limitations. CONCLUSIONS The challenges of operating at the cervicothoracic junction are widely known; yet, the selection of the most appropriate caudal level (C6 or cranial, C7, T1, T2 or caudal) for a multilevel PCDF has yet to be elucidated. This study suggests that the termination point for a PCDF may be less important than SVA axis alignment and other radiographic parameters in terms of overall patient outcomes. The revision rate and HRQOL outcomes correlate to SVA alignment and T1 slope, but it is irrespective of where the fusion ends. The results of this study suggest that patients have a higher postoperative NDI score with respect to cervical kyphosis and SVA misalignment. However, generalizations may not be true for all patients, and each operation must be tailored for each patient individually. We encourage future prospective studies that can shed further light on this topic.

reoperation after posterior cervical fusion that terminates cranial or caudal to the cervicalthoracic junction. Presented at: The Cervical Spine Research Society 42nd Annual Meeting. 2014. December 6, 2014; Orlando, FL.

alignment, sagittal deformity, and clinical implications: A review. J Neurosurg Spine. 2013;19: 141-159. 14. Gilad R, Gandhi CD, Arginteanu MS, Moore FM, Steinberger A, Camins M. Uncorrected sagittal plane imbalance predisposes to symptomatic instrumentation failure. Spine. 2008;8:911-917.

8. Auerbach J, Cho W, Riew KD. Crossing the cervicothoracic junction in multilevel posterior cervical fusions reduces the rate of symptomatic adjacent segment breakdown. Presented as a poster at the 37th Annual Meeting of the Cervical Spine Research Society, 2009. December 6, 2014; Orlando, FL.

15. Hann S, Chalouhi N, Madineni R, Vaccaro AR, Albert TJ, Harrop J, et al. An algorithmic strategy for selecting a surgical approach in cervical deformity correction. Neurosurg Focus. 2014;36:E5.

3. An HS, Wise JJ, Xu R. Anatomy of the cervicothoracic junction: A study of cadaveric dissection, cryomicrotomy, and magnetic resonance imaging. J Spinal Disord. 1999;12:519-525.

9. Harrison DE, Harrison DD, Cailliet R, Troyanovich SJ, Janik TJ, Holland B. Cobb method or Harrison posterior tangent method: Which to choose for lateral cervical radiographic analysis. Spine. 2000;25:2072-2078.

16. Juul T, Sogaard K, Davis AM, Roos EM. Psychometric properties of the neck outcome score, neck disability index, and short form-36 were evaluated in patients with neck pain. J Clin Epidemiol. 2016; 79:31-40.

4. Yang JS, Buchowski JM, Verma V. Construct type and risk factors for pseudarthrosis at the cervicothoracic junction. Spine. 2015;40:E613-E617.

10. Ames CP, Blondel B, Scheer JK, Schwab FJ, Le Huec JC, Massicotte EM, et al. Cervical radiographic alignment. Spine. 2013;38:149-160.

5. Schroeder GD, Kepler CK, Kurd MF, Mead L, Millhouse PW, Kumar P, et al. Is it necessary to extend a multilevel posterior cervical decompression and fusion to the upper thoracic. Spine. 2016; 41:1845-1849.

11. Kong L, Cao J, Wang L, Shen Y. Prevalence of adjacent segment disease following cervical spine surgery: A PRISMA-compliant systematic review and meta-analysis. Medicine. 2016;95:e4171.

6. Truumees E, Singh D, Geck M, Stokes J. Should long segment cervical fusions be routinely carried into the thoracic spine? Multi-center analysis. Spine J. 2017;17:43-44.

12. Passias PG, Oh C, Jalai CM, Worley N, Lafage R, Scheer JK, et al. Predictive model for cervical alignment and malalignment following surgical correction of adult spinal deformity. Spine. 2016; 41:E1096-E1103.

7. Wagner PJ, Adams S, Connolly PJ, et al. The incidence of instrumentation failure and

13. Scheer JK, Tang JA, Smith JS, Acosta FL Jr, Protopsaltis TS, Blondel B, et al. Cervical spine

2. Sekhon LH. Posterior cervical decompression and fusion for circumferential spondylotic cervical stenosis: Review of 50 consecutive cases. J Clin Neurosci. 2006;13:23-30.

e6

www.SCIENCEDIRECT.com

17. Hyun SJ, Kim KJ, Jahng TA, Kim HJ. Relationship between T1 slope and cervical alignment following multilevel posterior cervical fusion surgery: Impact of T1 slope minus cervical lordosis. Spine. 2016;41: E396-E402. 18. Tang JA, Scheer JK, Smith JS, Deviren V, Bess S, Hart RA, et al. The impact of standing regional cervical sagittal alignment on outcomes in posterior cervical fusion surgery. Neurosurgery. 2012;71: 662-669 [discussion: 669]. 19. Coe JD, Arlet V, Donaldson W, Berven S, Hanson DS, Mudiyam R, et al. Complications in spinal fusion for adolescent idiopathic scoliosis in the new millennium. Spine. 2006;31:345-349.

WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.10.062

ORIGINAL ARTICLE BROOKE T. KENNAMER ET AL.

CERVICAL ALIGNMENT IS CRUCIAL IN POSTERIOR CERVICOTHORACIC FUSION

20. Ondra SL. Biomechanics of spinal deformity correction. Clin Neurosurg. 2004;51:30-35.

for multilevel cervical myelopathy. Spine. 2006;31: 1568-1574.

21. Vaccaro AR, Lehman RA Jr, Hurlbert RJ, Anderson PA, Harris M, Hedlund R, et al. A new classification of thoracolumbar injuries: The importance of injury morphology, the integrity of the posterior ligamentous complex, and neurologic status. Spine. 2005;30:2325-2333.

23. Leven D, Cho S. Pseudarthrosis of the cervical spine: Risk factors, diagnosis, and management. Asian Spine J. 2016;10:776-786.

22. Ikenaga M, Shikata J, Tanaka C. Long-term results over 10 years of anterior corpectomy and fusion

Conflict of interest statement: The authors declare that the article content was composed in the absence of any

commercial or financial relationships that could be construed as a potential conflict of interest. Received 16 July 2018; accepted 8 October 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.10.062 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com

WORLD NEUROSURGERY -: e1-e7, - 2018

1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.

www.WORLDNEUROSURGERY.org

e7